Quick question.  All other things being equal, if you have a SIPS roof (8" thick) with integrated splines every 2 feet (4' wide panels) that are the entire 8" thick, compared with a roof that only uses non-structural splines every 4 feet, at the panel joints, how much strength does having the structural splines every 2 feet add?

rfleege;

great question! I suspect that they would not span the distance with out them

The roof strength requirement is based on the maximum displacement/deflection in the panel. For a cathedral ceiling (inside side of panel plastered) this is L/360 (L is the horizontal length of panel between supports). For an unfinshed interior side of panel (ie attic) it is L/240 (per Florida Building Code). On a simple gable roof maximum deflection takes place about halfway between between supports (ridge beam and supporting wall). Uniformly doubling your splines doubles the roof stregth from the splines but this does not double your panel strength since part of the panel strength is from the panel itself (primarily the skins).

The maximum deflection can be computed by treating the panel as a simply supported, uniformly loaded beam. The formula is:

maximum deflection = 5*w*L^4/(384EI). In this case the formula is more specifically:

maximum deflection = 5 * uniform load per horizontal length of panel * horizontal length of panel ^ 4 / (384 * (Epanel * Ipanel + Esplines * Isplines))

The uniform load is based on a combination of dead load (weight of panels and roofing material, account for extra weight if you might add solar), live load (12-20psf per FL code, depends on roof slope and size), wind loads, and snow loads (if your so lucky).
E is modulus of elasticity - The panel mfg should have a published value often in psi. This is essentially the strength of the material to resist bending/strain. You can also find published values for different materials.
I is moment of inertia - The panel mfg should have a published value often found through panel testing (US units are often inches^4). This is essentially a meaure of the cross sectional shapes ability to resist bending.
The moment of inertia for a simple rectangular cross-section spline would be: I = (width * height ^ 3) / 12. For I-joists the manufacturer will have published values.

You need to convert your units in the equation so you are using all inches or all feet. You can compare to published load span tables to check yourself. Either way you will probably need an engineer to check/sign off on your plans but this gives you an idea of how this effects your span capabilities. I would be careful about cutting it too close on allowable span lengths. A truss supported roof acts differently than a SIP roof. You do not want your SIP roof to have a noticable sag because you pushed the span capabilities.

I once replaced a roof that was designed for a 10 ft. span and it was actually spanning 20 ft.,  it did for about 2 years before it failed.

Posted By cmkavala on 29 Apr 2010 03:27 PM
I once replaced a roof that was designed for a 10 ft. span and it was actually spanning 20 ft.,  it did for about 2 years before it failed.

I didn't mean to imply that SIP roofs are more prone to failure.  In fact I think the opposite is true.  They should be able to more dynamically disperse impacts as long as the skin is not pierced.  I would think a metal panel with metal roof is ideal. 

The difference is a SIP roof panel can have 10 or more feet of unsupported panel allowing for some (preferably minor) long term creep/deflection in the middle of the span.  A truss roof seldom has more than a couple of feet of unsupported roof between trusses and individual chords along the truss itself. 

The limitations in span length are based on deflection which is much more stringent than expected failure (degree depends on material).  In addition it would be an unusual event to actually experience the design roof loads.  I would expect gross overspanning like you mentioned to be required to actaully have a roof fail.  I was just cautioning about pushing the span limit in terms of looks.  I would not want a noticable deflection in my roof to develop over time even if it was of no consequence structurally.

rfleege,

In my "guesstimation" a panel with intermediate internal 2x ribs @ 24" o.c. would span about 20% farther than a panel without internal ribs.

Here are some online span charts. They don't have your specific example, but you can get a sense of the strength increases using different splines.

Load Design Charts